Joty—On the Volume Change of Rocks and Minerals attending Fusion. 295 
bonate is formed and hung in the oven. When the melting point is reached, the 
adjustment of the current is refined till the bead appears to be in a fluctuating 
state. Now it is solid, now liquid. If these changes be very closely observed, it 
will be seen that they appear to originate at the centre of the exposed areas. But 
no adjustment was found sufficiently delicate to secure that the crystallization was 
for more than a moment confined to this central area. As the adjustment is capable 
of setting the current to the first place of decimals in the galvanometer readings 
(as plotted later in this Paper), and a decimal place at the slope of the curve 
at the point where potassium carbonate is located corresponds to a change in 
temperature of 10°, the inference is that the difference in temperature of the 
central areas from the side cannot be as much as this. For supposing that the 
difference was greater than this, supposing it was 20°, for example, then on 
adjusting the temperature of the oven so that the centre was just about to freeze, 
the sides will be 20° above the freezing point. Adjusting the temperature 
10° lower, the centre will now be frozen, but the sides will still be 10° above 
the freezing point. As a matter of observation this cannot be brought about. 
Lowering the temperature 10° when the centre is about to freeze insures the 
freezing of the whole globule. On these grounds it is thought that the error 
arising from this source cannot be considerable or at least comparable with the 
order of accuracy sought to be attained. 
A source of error which more obviously suggests itself is that attending changes 
of surface tension. The surface tension may be assumed to diminish as the fluidity 
gets greater, that is as the temperature rises; the amount of the change appearing 
to vary greatly in different substances. There are many substances which do not 
possess sufficient surface tension to permit of their treatment by this method. ‘The 
mineral apatite almost immediately on melting runs up the wire, wetting it like oil. 
At high temperatures olivine behaves in a similar manner. Basalt at the highest 
temperatures may also climb the wire. Silver sulphate reveals remarkable effects 
apparently due to marked enfeeblement of surface tension, at high temperatures. 
Beads of this substance flatten out somewhat, turning the flat faces towards the 
hot platinum. The bead in fact owes this distortion to a different distribution of 
surface tension over its surface. This substance when melted on platinum, runs 
over it like a thin oil. But observations on a great variety of salts and minerals 
show images which retain their spherical form till temperatures above the melting 
point. Thus orthoclase, according to observations on the meldometer, melts at 
1175° C. It was found practicable to carry measurements of its expansion up to 
about 1400°C. This substance remains viscous probably even at this temperature ; 
it indeed softens at so low a temperature as 865° C. No change of sphericity can 
be perceived. The basalt experimented with melted on the meldometer at 1153°, 
and runs about rapidly at 1173°. Perfect retention of form was shown till nearly 
